Respiratory diseases are a major cause of disability and death, and are associated with a significant healthcare burden.1 Moreover, the prevalence of these diseases continues to increase, spurred on by factors such as aging of the population, exposure to tobacco smoke, climatic and environmental conditions, and increasing obesity.2 In this context, the collection of objective data on the burden of care imposed by these diseases has become a priority for our National Health System (SNS, in its Spanish acronym), and one of the best sources of this information is hospital admissions.

In 1987, the Spanish Interterritorial SNS Council approved the implementation of the Minimum Basic Data Set (MBDS), an administrative database that hospitals in our health system are mandatorily required to complete. Since 2005, it also has been extended to the private sector, so the MBDS now contains more than 90% of discharges from acute care hospitals in Spain.3,4 It is therefore a very useful source for epidemiological research into respiratory diseases, and a resource with high organizational value for the planning and evaluation of health services.5

The MBDS database is completed by specially trained coding staff working in the admissions departments of Spanish hospitals. These professionals receive periodic refresher courses sponsored by their autonomous communities to ensure that they are properly trained.6 The coders use the medical discharge report and information from the hospital database to populate the MBDS database fields. In this way, they collect data such as age, sex, comorbidities, principal and secondary diagnoses, procedures conducted (both diagnostic and therapeutic), complications, mean length of stay, in-hospital mortality, destination at discharge, and re-admission within 30 days. It is important to note that each MBDS episode is assigned to a Diagnostic Related Group (DRG), a system that classifies hospital patients in homogeneous groups in terms of resource consumption.7 In order to ensure the quality of the coding, the autonomous communities also perform periodic audits.6 The data are sent annually to the Ministry of Health and Consumer Affairs, which is responsible for the management of the database.7 Some authors have studied the validity of the MBDS for different diseases and have corroborated its high sensitivity and specificity.8–10

The MBDS is, therefore, the largest administrative and clinical database available in the Spanish health system. Given its large sample size, analysis of this data could be considered as research known as big data. It is therefore a valuable tool for evaluating the epidemiology of respiratory diseases (morbidity), the management of health resources (in-hospital mortality, mean length of stay, readmissions, transfer to other social welfare centers) and patient safety (adverse reactions, complications associated with diagnostic and therapeutic procedures). Because the information is collected on a yearly basis, it can be analyzed to estimate the course over time of variables such as age, comorbidity, and diseases that have been treated over a specified time period.7

The results of the RECALAR project (Resources and Quality in Pathology of the Respiratory System) were recently published.11 By analyzing the MBDS, the authors showed that in 2015 respiratory diseases represented the second most frequent diagnosis, accounting for 12.1% of total discharges. Mean length of stay for respiratory diseases was 8.3 days. Respiratory diseases also played an important role in terms of mortality by causing 19.6% of the total deaths in that year.

In relation to epidemiological changes over time, analysis of the MBDS has revealed a decrease in recent years in the incidence of hospital admissions for some diseases, such as COPD12 and asthma.13 In contrast, there has been an increase in the incidence of hospitalizations for other conditions, such as respiratory infections,14 pulmonary fibrosis,15 and pulmonary thromboembolism.16 This increase, however, has occurred in parallel with a reduction in the mean length of stay16 and in-hospital mortality associated with some of these processes12,14–16, suggesting that their management in Spain has improved over time. As regards malignancies, the rate of lung cancer has fallen among men and risen among women.17

The use of the MBDS for clinical research has several advantages. It is faster and cheaper than collecting primary data, and easy to obtain.18 In addition, the sample size is high, conferring a great statistical power to the analysis of some variables, such as in-hospital mortality. It also useful for performing subgroup analyses by sex, age, comorbidities, and for studying diseases or procedures with very low prevalence. The database methodology has also remained stable over many years and its external validity is high, since it includes more than 95% of admissions to public and private hospitals.7 Nevertheless, its quality depends on accuracy of diagnosis and procedure coding, a process usually conducted on the basis of discharge reports.18 Another limitation of this database is the lack of significant clinical information, such as disease severity and duration, laboratory results, and treatments administered. Nor does it follow up patients after discharge. Finally, the MBDS encodes admissions, so the same patient can be “repeated” in the database if they were admitted several times during the same year.7

Despite these limitations, the use of administrative databases to evaluate health service outcomes has been validated in comparison with clinical records.19 The MBDS may therefore be seen as a valuable tool for epidemiological research into respiratory diseases, with their inherent social and healthcare importance. In any case, we must not forget that a large part of the clinical knowledge is still hidden, submerged in the clinical records of patients and non-standardized information gathered by health professionals in a natural language, an issue that big data will have to address in the near future.5